作者位次 #5/5ACL 2026Poster· Natural Language Processing
SGT : Securing Open-Source LLM s Against Malicious Fine-tuning via Safety Guidance Trigger
Sunguk Shin · Fangzhao Wu · Byung-Jun Lee · Meeyoung Cha · Sungwon Park
摘要
Open-weight large language models (LLMs) enable broad customization, but also increase exposure to post-release misuse, including malicious fine-tuning (MFT). To mitigate this risk, many prior defenses aim to improve the robustness of open-weight models to MFT by constraining adversarial fine-tun...